Data Viz III: Geographic Mapping with ggplot2

Description

In this class, we will explore how to create geographic maps using ggplot2, sf, maps, leaflet, and geojson. We’ll cover techniques for plotting data points on US and Canada maps, customizing map aesthetics, and working with spatial data.

1 Learning Objectives

By the end of this session, you will: 1. Load and visualize geographic data using sf and maps. 2. Create US and Canada maps using ggplot2. 3. Overlay data points on maps using geographic coordinates. 4. Customize map aesthetics for professional presentations. 5. Work with shapefiles and GeoJSON for more detailed geographic visualizations. 6. Implement a function for plotting. 7. Create an interactive map using leaflet.

Required Packages

Ensure the following packages are installed:

library(pacman)
p_load(ggplot2, sf, maps, dplyr, 
       ggthemes, ggrepel, leaflet, geojsonio,
       ggspatial, ggpubr)

2 Introduction to Mapping with ggplot2 and maps

We will start by creating basic maps of the US and Canada using the maps package.

2.1 US Map with State Boundaries

# Load US map data
us_map <- map_data("state")

# Plot US map
us_plot <- 
  ggplot() +
  geom_polygon(data = us_map, aes(x = long, y = lat, group = group),
               fill = "grey90", color = "grey35") +
  coord_fixed(1.3) +
  labs(title = "Map of the United States") +
  theme_base()

us_plot

2.2 Canada Map with Provincial Boundaries

# Load Canada map data
canada_map <- map_data("world", region = "Canada")

# Plot Canada map
canada_plot <- ggplot() +
  geom_polygon(data = canada_map, aes(x = long, y = lat, group = group),
               fill = "steelblue", color = "black") +
  coord_fixed(1.3) +
  labs(title = "Map of Canada")

canada_plot

canada_cut <- canada_plot +
  # Cut limits of map
  scale_y_continuous(limits = c(41, 48))+
  scale_x_continuous(limits = c(-87, -75))+
  theme_minimal()

canada_cut

3 Plotting Data Points on Maps

Next, we’ll plot specific data points (e.g., cities) on the US and Canada maps.

3.1 Plotting Cities on US Map

# Sample city data
cities_us <- data.frame(
  city = c("New Jersey", "New York", "Los Angeles", "Chicago", "Houston", "Phoenix"),
  lon = c(-74, -74.006, -118.2437, -87.6298, -95.3698, -112.074),
  lat = c(40.9, 40.7128, 34.0522, 41.8781, 29.7604, 33.4484)
)

# Plot US map with cities
us_plot +
  geom_point(data = cities_us, aes(x = lon, y = lat), 
             color = "red", size = 3) +
  # geom_label(data = cities_us, aes(x = lon, y = lat, label = city), 
  #                  size = 3, color = "black")
  geom_label_repel(data = cities_us, aes(x = lon, y = lat, label = city), 
                    size = 3, color = "black")

3.2 Plotting Cities on Canada Map

# Sample city data for Canada
cities_canada <- data.frame(
  city = c("Toronto", "Vancouver", "Montreal", "Calgary", "Ottawa"),
  lon = c(-79.3832, -123.1216, -73.5673, -114.0719, -75.6972),
  lat = c(43.6511, 49.2827, 45.5017, 51.0447, 45.4215)
)

# Plot Canada map with cities
canada_plot +
  geom_point(data = cities_canada, aes(x = lon, y = lat), 
             color = "blue", size = 3) +
  geom_label_repel(data = cities_canada, aes(x = lon, y = lat, label = city), 
                   size = 3, color = "black")

4 Mapping with Shapefiles using sf

For more detailed geographic visualizations, we can use shapefiles with the sf package.

4.1 Loading and Plotting Shapefiles

# Load shapefile (replace 'path_to_shapefile' with your actual path)
# usa_shapefile <- st_read("path_to_shapefile/usa_shapefile.shp")

# Example using built-in dataset from `sf`
nc <- st_read(system.file("shape/nc.shp", package = "sf"))

Reading layer `nc' from data source 
  `/Library/Frameworks/R.framework/Versions/4.4-arm64/Resources/library/sf/shape/nc.shp' 
  using driver `ESRI Shapefile'
Simple feature collection with 100 features and 14 fields
Geometry type: MULTIPOLYGON
Dimension:     XY
Bounding box:  xmin: -84.32385 ymin: 33.88199 xmax: -75.45698 ymax: 36.58965
Geodetic CRS:  NAD27

# Plot shapefile
ggplot(nc) +
  geom_sf(fill = "lightblue", color = "black") +
  labs(title = "Shapefile Example: North Carolina Counties") +
  # Adjust scales for limits
  scale_y_continuous(limits = c(34.5, 36))+
  scale_x_continuous(limits = c(-82, -78))+
  theme_minimal()

4.2 Read shp of Canada and US

## Load shp files
usa_shp <- st_read("shp_map/us/usa.shp") %>%
  # Remove non-contiguous and territories
  filter(!(NAME %in% c("Alaska", "District of Columbia", "Hawaii", "Puerto Rico")))

Reading layer `usa' from data source 
  `/Users/acorrend/Documents/GitHub/plnt6800/coding/week_05/shp_map/us/usa.shp' 
  using driver `ESRI Shapefile'
Simple feature collection with 52 features and 9 fields
Geometry type: MULTIPOLYGON
Dimension:     XY
Bounding box:  xmin: -179.1473 ymin: 17.88481 xmax: 179.7785 ymax: 71.35256
Geodetic CRS:  NAD83

can_shp <- st_read("shp_map/can/canada.shp")

Reading layer `canada' from data source 
  `/Users/acorrend/Documents/GitHub/plnt6800/coding/week_05/shp_map/can/canada.shp' 
  using driver `ESRI Shapefile'
Simple feature collection with 13 features and 6 fields
Geometry type: MULTIPOLYGON
Dimension:     XY
Bounding box:  xmin: -141.0181 ymin: 41.67695 xmax: -52.5823 ymax: 89.99943
Geodetic CRS:  NAD83

4.3 Create objects for maps

# Create list of selected provinces
selected_provinces <- c("Ontario", "Manitoba", "Quebec")

# Define coordinates for Ontario cities
ontario_cities <- data.frame(
  city = c("Toronto", "Ottawa", "Hamilton", "London", "Kingston"),
  lon = c(-79.3832, -75.6972, -79.8711, -81.2497, -76.4880),
  lat = c(43.6511, 45.4215, 43.2557, 42.9834, 44.2312)
)

4.4 Define function to customize plot

geo_plot <- function(x, y, z, title = NULL){
  ggplot()+
    geom_sf(data=x, fill = "white", color = "black") + # Provinces map
  geom_sf(data=y, fill = "white", color = "black")+ # US map
  # Adjust scales for lat and lon
  scale_y_continuous(limits = c(41.8, 46))+
  scale_x_continuous(limits = c(-84, -75), breaks = seq(-84,-74, by=1)) +
  # Add cities with points
  geom_point(data = z, aes(x = lon, y = lat, fill = city), 
             color = "grey25", shape = 21, size = 3, alpha = 0.95) +
  # Scalebar
  annotation_scale(tick_height = 0.3)+
  # Text Notes for names of cities
  geom_text_repel(data = z, 
                  aes(x=lon, y=lat, label = city), size = 3)+
  # Name of PROVINCE
  annotate("text", x = -78, y = 45, label = "ONTARIO", 
           size = 4, fontface = "bold")+
  # Name of Lakes
  ## Ontario
  annotate("text", x = -77.8, y = 43.7, label = "Lake Ontario", 
           size = 3, fontface = "italic")+
  ## Huron 
  annotate("text", x = -82.5, y = 44.5, label = "Lake \nHuron", 
           size = 3, fontface = "italic")+
  # Add labels
  labs(title = title, 
       x = "Longitude", y = "Latitude")+
  # Adjust theme
  theme_base()+
  # reduce axis text size
  theme(
    panel.background = element_rect(fill = "#bde0fe"),
    title = element_text(size = rel(.7)),
    axis.title =  element_text(size = rel(.9), face = "bold"),
    axis.text = element_text(size = rel(.5))
    )
}

4.5 Plot

# Plot from function
geo_plot(x = can_shp, y = usa_shp, z = ontario_cities, 
         title = "Ontario Map from SHP")

5 Working with GeoJSON Data

We can use GeoJSON files for geographic data. Here’s how to load and visualize a GeoJSON file for Ontario, Canada.

5.1 Loading GeoJSON Data

# Load Canada GeoJSON (replace with actual file path if available)
# Read geojson for Canada
can_geojson <- read_sf("geojson_maps/canada_provinces.geojson")  %>%  
  # dplyr::filter(name == "Ontario")  %>% 
  dplyr::filter(name %in% selected_provinces)  %>% 
  dplyr::mutate(Province = name, 
                GEOID = cartodb_id)  %>%  
  dplyr::select(GEOID, Province, geometry)

# Read geojson for US
usa_geojson <- read_sf("geojson_maps/us-states.json")

5.2 Plot GEOJSON

# Plot from function
geo_plot(x = can_geojson, y = usa_geojson, z = ontario_cities, 
         title = "Ontario Map from GEOJSON")

5.3 Explanation:

• geojson_read(): Reads the GeoJSON file.
• st_as_sf(): Converts the data into an sf object for plotting.
• geom_sf(): Plots the GeoJSON data.

6 Creating Interactive Maps with leaflet

leaflet allows us to create interactive maps. Let’s create a map of Ontario, Canada.

6.1 Ontario Map with leaflet

# Create interactive map
ontario_cities %>%
leaflet() %>%
  addTiles() %>%
# Add PINS
addMarkers(~lon, ~lat, popup = ~city) %>%
addCircleMarkers(~lon, ~lat, popup = ~city, radius = 5, color = "gold",
                  fillOpacity = 0.7) %>%
# Configure initial view of the map
setView(lng = -80, lat = 44, zoom = 6)

6.2 Explanation:

• addTiles(): Adds the base map layer.
• addMarkers(): Plots city locations with popups displaying city names.
• setView(): Centers the map on Toronto with a specified zoom level.

7 Conclusion

In this lesson, you learned how to: - Create geographic maps of the US and Canada using ggplot2 and maps. - Plot data points on maps with latitude and longitude coordinates. - Use shapefiles and GeoJSON for more detailed geographic visualizations with sf. - Customize map aesthetics for clearer and more professional presentations. - Create interactive maps using leaflet.

These techniques are valuable for visualizing spatial data and can be adapted for a wide range of research applications.

8 Additional Resources

1. ggplot2 Documentation – Official documentation for creating maps and more.
2. sf Package Documentation – Guide to handling spatial data in R.
3. R Maps Package – Reference for basic map data.
4. Leaflet for R – Documentation for interactive mapping in R.
5. GeoJSON Data and Tools – Create and explore GeoJSON data.
6. R Spatial Data Science – Tutorials and resources for spatial data analysis.
7. R Graph Gallery - Maps – Examples of different map types using R.

Experiment with different datasets and explore additional map customizations to further enhance your visualizations! 🚀